OBJECTIVE

To evaluate glycemic outcomes in youth (aged 13–25 years) with type 1 diabetes and high-risk glycemic control (HbA1c ≥8.5% [69 mmol/mol]) on multiple daily injection (MDI) therapy after transitioning to advanced hybrid closed loop (AHCL) therapy.

RESEARCH DESIGN AND METHODS

This prospective, 3-month, single-arm, dual-center study enrolled 20 participants, and all completed the study.

RESULTS

HbA1c decreased from 10.5 ± 2.1% (91.2 ± 22.8 mmol/mol) at baseline to 7.6 ± 1.1% (59.7 ± 11.9 mmol/mol), and time spent in target range 70–180 mg/dL (3.9–10.0 mmol/L) increased from 27.6 ± 13.2% at baseline to 66.5 ± 9.8% after 3 months of AHCL. Two episodes of diabetic ketoacidosis attributed to infusion set failure occurred.

CONCLUSIONS

AHCL has the potential to improve suboptimal glycemia in youth with type 1 diabetes previously on MDI therapy.

Youth with type 1 diabetes are at high risk for suboptimal glycemic control and low treatment adherence (1,2). As HbA1c during youth is highly predictive of long-term HbA1c trajectory, interventions are urgently required to alter a life course predictive for premature development of diabetes complications (3). To our knowledge, the only study using an advanced hybrid closed loop (AHCL) system to target high-risk youth previously on multiple daily injection (MDI) therapy enrolled a younger cohort (aged 7–17 years) and excluded those with a very high HbA1c of >12.4% (112 mmol/mol) (4). Our objective was to explore the impact of AHCL in youth on MDI therapy, focusing on those with high-risk glycemic control by removing any upper cutoff exclusion criteria.

This prospective, single-arm, dual-center study investigated AHCL in 20 youth with high-risk glycemia on MDI therapy. The study was conducted in compliance with all applicable regulatory requirements, approved by the Southern Health and Disability Ethics Committee (Wellington, New Zealand) (21/STH/33), and registered with the Australian New Zealand Clinical Trials Registry. Participants were recruited from Dunedin and Christchurch (New Zealand) public hospitals on a first-come, first-served basis. Eligibility criteria were type 1 diabetes as per American Diabetes Association classification for ≥1 year (5), age 13–25 years (inclusive), current HbA1c of ≥8.5% (69 mmol/mol), on MDI therapy, and daily insulin ≥8 units. Exclusion criteria were pregnancy, medication indicative of diabetes complications (ACE inhibitors and statins permitted), systemic glucocorticoids, sodium–glucose cotransporter 2 inhibitors, glucagon-like peptide 1 receptor agonists, hydroxyurea, and any concomitant condition that would interfere with the study conduct or pose an unacceptable risk to participants. Prior to study procedures, appropriate informed consent/assent was obtained. AHCL study data by our group in patients aged 14–21 years with well-controlled glycemia showed improvements in time in range (TIR) of 14.4% (6). Therefore, to estimate mean changes in TIR to a 95% precision interval of ±4%, 20 participants were required.

Procedures

Enrollment occurred between 6 October 2021 and 1 February 2022. Following a 2-week baseline data collection using blinded continuous glucose monitoring (CGM) (Guardian Sensor 3 with Guardian Link 3 transmitter; Medtronic, San Francisco, CA), participants underwent training and used the MiniMed 780G AHCL system (Medtronic) for 3 months. We adapted a 10-day initiation protocol for our study (4), the key adaptation being a structured 72-h rapid onboarding (protocol described in the Supplementary Methods).

This trial used an AHCL system consisting of the MiniMed 780G insulin pump with the Guardian Sensor 3 CGM and Link 3 transmitter. Further details about the 780G SmartGuard feature are described in the Supplementary Methods.

Outcomes

The coprimary outcomes were changes in TIR 70–180 mg/dL (3.9–10.0 mmol/L) and HbA1c between baseline and after AHCL intervention. HbA1c was measured by a point-of-care device (DCA Vantage Analyzer; Siemens Healthcare Diagnostics, Swords, Ireland). However, if HbA1c at screening was outside the detection range of the point-of-care device (2.5–14% [4–130 mmol/mol]), all HbA1c tests for that participant were performed by a diagnostic laboratory. Secondary outcomes included time below range, time above range, TIR differentiated by day (0600–2359 h) and night (0000–0559 h), coefficient of variation (CV), and AHCL system characteristics. Comparisons between glycemic outcomes were made using the 14-day average of baseline, 3-day average of sensor-augmented pump with predictive low-glucose management, 14-day average of 4–6 weeks AHCL, and 14-day average of 11–13 weeks of AHCL data. Serious adverse events were defined as severe hypoglycemia and diabetic ketoacidosis (DKA).

Statistical Analysis

CGM and insulin data were collected from CareLink software. Mean changes in outcomes were estimated at each time point in the study after baseline along with the 95% CI. Statistical analyses were performed using Stata 17.0 software (StataCorp, College Station, TX).

All 20 participants completed the study. Baseline participant demographics are shown in Table 1. BMI was 25.2 ± 5.9 kg/m2 at baseline and 26.6 ± 6.3 kg/m2 after the 3-month intervention.

Table 1

Baseline demographic and clinical characteristics

Participants (n = 20)
Age (years), mean (range) 18.8 (13.3–25.7) 
Sex, n (%)  
 Female 12 (60) 
 Male 8 (40) 
Prioritized ethnicity, n (%)  
 New Zealand European 17 (85) 
 Māori* 1 (5) 
 Samoan 1 (5) 
 Other 1 (5) 
Socioeconomic deprivation index,n (%)  
 Low (1–3) 7 (35) 
 Medium (4–7) 8 (40) 
 High (8–10) 5 (25) 
Duration of diabetes (years), mean ± SD 9.7 ± 5.4 
CGM use prior to enrollment, n (%)  
Intermittently scanned CGM (FreeStyle Libre) 10 (50) 
Real-time CGM (Dexcom G6) 1 (5) 
TDD (units/day), mean ± SD 71.1 ± 25.7 
TDD (units/kg/day), mean ± SD 1.0 ± 0.3 
BMI (kg/m2), mean ± SD 25.2 ± 5.9 
HbA1c (%), mean ± SD [range] 10.5 ± 2.1 [8.6–17.6] 
HbA1c (mmol/mol), mean ± SD [range] 91.2 ± 22.8 [70–169] 
Participants (n = 20)
Age (years), mean (range) 18.8 (13.3–25.7) 
Sex, n (%)  
 Female 12 (60) 
 Male 8 (40) 
Prioritized ethnicity, n (%)  
 New Zealand European 17 (85) 
 Māori* 1 (5) 
 Samoan 1 (5) 
 Other 1 (5) 
Socioeconomic deprivation index,n (%)  
 Low (1–3) 7 (35) 
 Medium (4–7) 8 (40) 
 High (8–10) 5 (25) 
Duration of diabetes (years), mean ± SD 9.7 ± 5.4 
CGM use prior to enrollment, n (%)  
Intermittently scanned CGM (FreeStyle Libre) 10 (50) 
Real-time CGM (Dexcom G6) 1 (5) 
TDD (units/day), mean ± SD 71.1 ± 25.7 
TDD (units/kg/day), mean ± SD 1.0 ± 0.3 
BMI (kg/m2), mean ± SD 25.2 ± 5.9 
HbA1c (%), mean ± SD [range] 10.5 ± 2.1 [8.6–17.6] 
HbA1c (mmol/mol), mean ± SD [range] 91.2 ± 22.8 [70–169] 
*

Māori are the indigenous people of New Zealand.

New Zealand index of socioeconomic deprivation 2018 in which 1 represents the least and 10 the most socioeconomic deprivation (20).

Accuracy of this self-reported figure is uncertain, considering the suboptimal baseline HbA1c.

Glycemic Outcomes

Mean HbA1c decreased by 2.9 percentage points (31.5 mmol/mol) from 10.5 ± 2.1% (91.2 ± 22.8 mmol/mol) at baseline to 7.6 ± 1.1% (59.7 ± 11.9 mmol/mol) after AHCL (Fig. 1 and Supplementary Table 1). TIR 70–180 mg/dL (3.9–10.0 mmol/L) increased by 38.9 percentage points (95% CI 31.2, 46.5) from 27.6 ± 13.2% at baseline to 66.5 ± 9.8% after AHCL (Fig. 2, Supplementary Fig. 1, and Supplementary Table 1). Mean glucose and glucose variability were also lower at 3 months (sensor glucose change −83 mg/dL [95% CI −106, −61], CV change −4.3 percentage points [−8.6, −0.1]). Further secondary outcomes are presented in Fig. 2 and Supplementary Table 1.

Figure 1

Changes in HbA1c after 3 months of AHCL use. Data are shown as individual values (blue circles) and mean ± SD of individual values (red squares). Minimal individual HbA1c reduction was 1.7% (18 mmol/mol), and maximal individual HbA1c reduction was 9.7% (106 mmol/mol).

Figure 1

Changes in HbA1c after 3 months of AHCL use. Data are shown as individual values (blue circles) and mean ± SD of individual values (red squares). Minimal individual HbA1c reduction was 1.7% (18 mmol/mol), and maximal individual HbA1c reduction was 9.7% (106 mmol/mol).

Close modal
Figure 2

Glycemic outcomes after baseline, sensor-augmented pump with predictive low-glucose management (SAP+PLGM), 6 weeks AHCL, and 3 months AHCL. Data are mean ± SD or mean and represent the 14-day average of baseline, 4–6 weeks of AHCL use, 11–13 weeks of AHCL data, and the 3-day average of SAP+PLGM data. SG, sensor glucose.

Figure 2

Glycemic outcomes after baseline, sensor-augmented pump with predictive low-glucose management (SAP+PLGM), 6 weeks AHCL, and 3 months AHCL. Data are mean ± SD or mean and represent the 14-day average of baseline, 4–6 weeks of AHCL use, 11–13 weeks of AHCL data, and the 3-day average of SAP+PLGM data. SG, sensor glucose.

Close modal

AHCL Performance

System settings, performance, and insulin delivery after 3 months are presented in Table 2. SmartGuard was active 91 ± 11.2% of time, with sensor wear during 86.3 ± 12.2% of the time. Mean insulin total daily dose (TDD) was 72.1 ± 38.9 units, of which 47.1 ± 6.5% were administered autobasally. Of the remaining total bolus insulin, 51.2 ± 17.2% of units were administered through autocorrection. On average, participants entered 119.9 ± 64.8 g of carbohydrates across 2.8 ± 1.4 meals into the system per day.

Table 2

System performance and insulin delivery distribution at end of the AHCL intervention

Participants (n = 20)
Sensor glucose target, n (%)  
 100 mg/dL (5.5 mmol/L) 17 (85) 
 110 mg/dL (6.1 mmol/L) 2 (10) 
 120 mg/dL (6.7 mmol/L) 1 (5) 
AIT, h (n2.0 (19); 2.5 (1) 
Time in AHCL (%) 91.0 ± 11.2 
Sensor wear (%) 86.3 ± 12.2 
SMBG per day (n2.6 ± 0.5 
Meals per day (n2.8 ± 1.4 
Carbohydrates per day (g) 119.9 ± 64.8 
AHCL exits per week (n2.7 ± 2.0 
User-initiated AHCL exits per week (n0.4 ± 0.7 
System-initiated AHCL exits per week (n2.3 ± 1.9 
TDD (units/day) 72.1 ± 38.9 
TDD (units/kg/day) 0.9 ± 0.4 
Basal insulin (% of TDD) 47.1 ± 6.5 
Bolus insulin (% of TDD) 52.9 ± 6.5 
Autocorrections (% of bolus insulin) 51.2 ± 17.2 
Participants (n = 20)
Sensor glucose target, n (%)  
 100 mg/dL (5.5 mmol/L) 17 (85) 
 110 mg/dL (6.1 mmol/L) 2 (10) 
 120 mg/dL (6.7 mmol/L) 1 (5) 
AIT, h (n2.0 (19); 2.5 (1) 
Time in AHCL (%) 91.0 ± 11.2 
Sensor wear (%) 86.3 ± 12.2 
SMBG per day (n2.6 ± 0.5 
Meals per day (n2.8 ± 1.4 
Carbohydrates per day (g) 119.9 ± 64.8 
AHCL exits per week (n2.7 ± 2.0 
User-initiated AHCL exits per week (n0.4 ± 0.7 
System-initiated AHCL exits per week (n2.3 ± 1.9 
TDD (units/day) 72.1 ± 38.9 
TDD (units/kg/day) 0.9 ± 0.4 
Basal insulin (% of TDD) 47.1 ± 6.5 
Bolus insulin (% of TDD) 52.9 ± 6.5 
Autocorrections (% of bolus insulin) 51.2 ± 17.2 

Data are mean ± SD unless otherwise indicated. AIT, active insulin time; SMBG, self-monitored blood glucose.

Serious Adverse Events

No cases of severe hypoglycemia occurred during the study. Two episodes of mild to moderate DKA (pH 7.16 and 7.24) occurred in two participants (DKA rate of 0.4 per 1 patient-year), both likely due to infusion set occlusion. In comparison, five episodes of DKA were experienced by four participants within 12 months prior to the study (DKA rate of 0.25 per 1 patient-year). All adverse events were resolved without sequelae.

This prospective, single-arm, dual-center clinical trial describes the 3-month outcomes of MiniMed 780G AHCL use following a structured 72-h rapid onboarding protocol. We demonstrate that youth with type 1 diabetes and high-risk glycemic control previously using MDI therapy have the potential to improve glycemia using AHCL.

This study describes the greatest gains so far reported for youth with high-risk glycemia, with a mean HbA1c improvement of 2.9 percentage points (31.5 mmol/mol). Mean TIR improved by 38.9 percentage points, mostly accounted for by a reduction of 38.4 percentage points in time spent above 250 mg/dL (13.9 mmol/L). Comparison of results across different studies is challenging and should arguably be avoided because of differences in study design, population characteristics, and closed loop systems used. These limitations considered, outcomes of this study are in line with those reported for larger trials using various MiniMed AHCL systems (68), and studies using other advanced automated insulin delivery systems (Control-IQ, CamAPS FX, AndroidAPS) (911). One single-arm study reported an increase of 36.7 percentage points in TIR and reduction of 2.1 percentage points in HbA1c after MiniMed 780G AHCL use in a younger cohort with healthier, but nonetheless suboptimal, glycemic control (aged 7–17 years, mean baseline HbA1c 8.6%) (4).

Of the insulin TDD, 74% of units were automatically computed and delivered by the system without user input, reflecting the benefits of AHCL on treatment burden in this complex cohort. The system was successfully used, with mean sensor wear time of 86% and 91% of time spent in AHCL. Of the 2.7 AHCL exits occurring per week, 15% were user initiated. No participant attrition occurred. These outcomes are improved over previous first-generation hybrid closed loop outcomes (12,13), indicating high participant engagement and system acceptability likely attributed to further automation of the MiniMed 780G AHCL system and suggesting an emerging therapeutic paradigm shift from conventional therapy to automated insulin delivery systems.

Intriguingly, the greatest individual gains were experienced by a participant with baseline HbA1c of 17.6% (169 mmol/mol) and 0% TIR, improving to an HbA1c of 7.9% (63 mmol/mol) and TIR of 60% after 3 months. Moreover, with only 1.6% of TDD delivered as a user-initiated bolus, the system provided improved glycemic control despite missed meal boluses, emphasizing the observation of improved outcomes in all participants regardless of baseline HbA1c (14). These results reinforce that those with the highest risk glycemia have a potential for the greatest gains from AHCL with respect to reducing burden and risk of future complications (15).

Safety of AHCL technology is an important consideration in this population. No severe hypoglycemia occurred, in line with previous AHCL data (5,711). Two episodes of mild to moderate DKA occurred, with both due to infusion set failure/occlusion and contributed to by the novelty of pump therapy. Infusion set failure/occlusion is a well-documented complication of all insulin pump therapies, with higher rates seen in younger users (16). Therefore, frequent opportunistic anticipatory education to avoid and manage infusion set issues remains essential (17).

Limitations of this study are the small number of participants and lack of a comparator. The short 3-month follow-up period did not allow for analysis of long-term results, but participants will be followed for a further 9 months to confirm whether outcomes are sustained. Selection bias due to more motivated individuals participating presents another potential limitation. Of note is that of 23 individuals with high-risk glycemic control approached, 20 agreed to participate. Further studies with greater ethnic diversity would also be important. Overall, 55% of participants used CGM prior to enrollment. While CGM alone can lead to glycemic improvements, a recent systematic review and meta-analysis of randomized controlled trials demonstrated that CGM use alone accounts for only 5.4% TIR improvement (18). Strengths of this study are the structured training allowing for rapid onboarding, the even distribution of socioeconomic deprivation, and the high baseline HbA1c compared with other studies.

For less complex type 1 diabetes populations, closed loop systems are already the gold standard therapeutic option (19). AHCL, combined with adequate training and clinical support, should now be considered a first-line therapeutic tool for those with the most to gain, namely youth with high-risk glycemia.

Clinical trial reg. no. ACTRN12621000556842, www.anzctr.org.au

This article contains supplementary material online at https://doi.org/10.2337/figshare.21785207.

Acknowledgments. The authors thank the participants and their families for taking part in this study.

Funding. This study was funded by Lottery Health Research grant LHR-2021-153881.

Duality of Interest. The diabetes technology used in this study was provided by Medtronic. No other potential conflicts of interest relevant to this study were reported.

Author Contributions. A.B. conducted data analysis. A.B., A.S.W., C.M.F., O.J.S., S.D.J., P.J.M.-H., M.I.d.B., and B.J.W. researched data. A.B., J.J.H., and B.J.W. revised the manuscript. A.B., M.I.d.B., and B.J.W. wrote the manuscript. J.J.H. designed and conducted the statistical analysis. M.I.d.B. and B.J.W. conceptualized the study and acquired funding. All authors approved the final manuscript as submitted. B.J.W. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. Parts of this study were presented in invited oral presentations at the 12th Biennial Scientific Meeting of the Asia Pacific Paediatric Endocrine Society, Seoul, South Korea, 5–8 October 2022, and at the 48th Annual International Society for Pediatric and Adolescent Diabetes Conference, Abu Dhabi, United Arab Emirates, 13–16 October 2022.

1.
Foster
NC
,
Beck
RW
,
Miller
KM
, et al
.
State of type 1 diabetes management and outcomes from the T1D Exchange in 2016-2018
.
Diabetes Technol Ther
2019
;
21
:
66
72
2.
Datye
KA
,
Moore
DJ
,
Russell
WE
,
Jaser
SS
.
A review of adolescent adherence in type 1 diabetes and the untapped potential of diabetes providers to improve outcomes
.
Curr Diab Rep
2015
;
15
:
51
3.
Nirantharakumar
K
,
Mohammed
N
,
Toulis
KA
,
Thomas
GN
,
Narendran
P
.
Clinically meaningful and lasting HbA1c improvement rarely occurs after 5 years of type 1 diabetes: an argument for early, targeted and aggressive intervention following diagnosis
.
Diabetologia
2018
;
61
:
1064
1070
4.
Petrovski
G
,
Al Khalaf
F
,
Campbell
J
, et al
.
Glycemic outcomes of advanced hybrid closed loop system in children and adolescents with type 1 diabetes, previously treated with multiple daily injections (MiniMed 780G system in T1D individuals, previously treated with MDI)
.
BMC Endocr Disord
2022
;
22
:
80
5.
American Diabetes Association
.
2. Classification and diagnosis of diabetes: Standards of Medical Care in Diabetes-2021
.
Diabetes Care
2021
;
44
(
Suppl. 1
):
S15
S33
6.
Collyns
OJ
,
Meier
RA
,
Betts
ZL
, et al
.
Improved glycemic outcomes with Medtronic MiniMed advanced hybrid closed-loop delivery: results from a randomized crossover trial comparing automated insulin delivery with predictive low glucose suspend in people with type 1 diabetes
.
Diabetes Care
2021
;
44
:
969
975
7.
Bergenstal
RM
,
Nimri
R
,
Beck
RW
, et al.;
FLAIR Study Group
.
A comparison of two hybrid closed-loop systems in adolescents and young adults with type 1 diabetes (FLAIR): a multicentre, randomised, crossover trial
.
Lancet
2021
;
397
:
208
219
8.
Choudhary
P
,
Kolassa
R
,
Keuthage
W
, et al.;
ADAPT Study Group
(2022)
.
Advanced hybrid closed loop therapy versus conventional treatment in adults with type 1 diabetes (ADAPT): a randomised controlled study
.
Lancet Diabetes Endocrinol
2022
;
10
:
720
731
9.
Brown
SA
,
Kovatchev
BP
,
Raghinaru
D
, et al.;
iDCL Trial Research Group
.
Six-month randomized, multicenter trial of closed-loop control in type 1 diabetes
.
N Engl J Med
2019
;
381
:
1707
1717
10.
Tauschmann
M
,
Thabit
H
,
Bally
L
, et al.;
APCam11 Consortium
.
Closed-loop insulin delivery in suboptimally controlled type 1 diabetes: a multicentre, 12-week randomised trial
.
Lancet
2018
;
392
:
1321
1329
11.
Burnside
MJ
,
Lewis
DM
,
Crocket
HR
, et al
.
Open-source automated insulin delivery in type 1 diabetes
.
N Engl J Med
2022
;
387
:
869
881
12.
Messer
LH
,
Berget
C
,
Vigers
T
, et al
.
Real world hybrid closed-loop discontinuation: predictors and perceptions of youth discontinuing the 670G system in the first 6 months
.
Pediatr Diabetes
2020
;
21
:
319
327
13.
Berget
C
,
Messer
LH
,
Vigers
T
, et al
.
Six months of hybrid closed loop in the real-world: an evaluation of children and young adults using the 670G system
.
Pediatr Diabetes
2020
;
21
:
310
318
14.
Ekhlaspour
L
,
Town
M
,
Raghinaru
D
,
Lum
JW
,
Brown
SA
,
Buckingham
BA
.
Glycemic outcomes in baseline hemoglobin A1c subgroups in the International Diabetes Closed-Loop Trial
.
Diabetes Technol Ther
2022
;
24
:
588
591
15.
Nathan
DM
,
Genuth
S
,
Lachin
J
, et al.;
Diabetes Control and Complications Trial Research Group
.
The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus
.
N Engl J Med
1993
;
329
:
977
986
16.
Ross
PL
,
Milburn
J
,
Reith
DM
,
Wiltshire
E
,
Wheeler
BJ
.
Clinical review: insulin pump-associated adverse events in adults and children
.
Acta Diabetol
2015
;
52
:
1017
1024
17.
Milburn
J
,
de Lange
M
,
Wiltshire
E
, et al
.
Evaluating the impact of an insulin pump discontinuation action plan on patient or caregiver confidence and anxiety
.
J Diabetes Metab Disord
2019
;
18
:
81
88
18.
Elbalshy
M
,
Haszard
J
,
Smith
H
, et al
.
Effect of divergent continuous glucose monitoring technologies on glycaemic control in type 1 diabetes mellitus: a systematic review and meta-analysis of randomised controlled trials
.
Diabet Med
2022
;
39
:
e14854
19.
Boughton
CK
,
Hovorka
R
.
The artificial pancreas
.
Curr Opin Organ Transplant
2020
;
25
:
336
342
20.
Atkinson
J
,
Salmond
C
,
Crampton
P
.
NZDep2018 Index of Deprivation: Interim Research Report, December 2019
.
Accessed 2 August 2022. Available from https://www.otago.ac.nz/wellington/otago730394.pdf
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at https://www.diabetesjournals.org/journals/pages/license.